This is probably the most wrongly treated setting parameter. In fact, it’s all quite simple – when the audio engineer sets ratio value, he defines how much the audio which overruns the threshold will be compressed, in other words, literally how many times quieter these peaks will be after processing. So ratio 1:1 means that the signal won’t experience any changes. If the ratio makes up 2:1, the peaks at the output will be twice quieter. For example, the signal that has overrun the preset threshold by 8 dB will be attenuated by 4 dB. In practice of professional mixing and mastering, ratio values of 2:1–3:1 are considered to be moderate and used for the most possible unnoticeable intervention in sound dynamics. Value 5:1 is already the compression of medium hardness (it’s frequently used for drum mixing; drums’ character is explosive and dynamic by nature). 8:1 is a fairly powerful compression with an evidently audible effect. The values from 20:1 and higher up to ∞:1 (infinity to one) mean that the compressor is already working as a limiter by limiting peaks hard without letting anything higher than the preset threshold to come past. It’s often used in professional music mixing for adding evidently distorted smashed sounding.
Although the role of compression in enhancement of perceived loudness is well-known, in fact, these manipulations break the natural structure of the recorded signal and make it sound a lot quieter… But here “Output Gain” and ‘Make-up Gain” are coming into play. Through this parameter, the mixing mastering engineer restores the loudness lost due to processing. As a matter of convenience many devices have “Gain Reduction” meters that allow engineers to track the work of a compressor during mixing and to use correct output gain values.